Absence, Substitutability and Productivity: Evidence from Teachers Asma Benhenda Paris School of Economics UC Berkeley – April 19, 2017 Still Preliminary 1 / 39
Motivation • Worker absence : frequent in many countries (2 to 3 % of annual work time in the US, UK & France) • Empirical evidence on the causal effect of worker absence on productivity is scarce (Clotfelter et al., 2009 ; Duflo et al., 2012 ; Herrmann and Rockoff, 2012) • Even much less is known on organizations’ strategies to cope with this disruptive event • I address this issue for teachers : • When a teacher is absent, how does it hurt student achievement ? • How easily do schools manage to mitigate this effect with substitute teachers ? 2 / 39
What this Paper Does • The aim of this paper is : 1 to estimate the effect of teacher absence on student test scores in 9th grade 2 to study how the effect of teacher absences is mitigated by the assignment of substitute teachers 3 to study how the effect of substitute teachers depend on their type (tenured substitutes vs. contract teachers) • Important questions because : • Impact of worker health and effort on productivity • Specific human capital and its relationship with worker substituability • Teachers : consequences for educational inequalities 3 / 39
Setting • Two types of absences : • Less than a year : not systematically replaced + disruptive for students • One year or more : systematically replaced + not disruptive for students ⇒ disentangle substitute teacher’s quality effect from other effects Two main types of substitute teachers : • • Tenured teachers assigned to a ZIP code area ( Titulaires sur Zone de Remplacement ) • Contract teachers hired on the spot, not trained nor certified 4 / 39
Data and Empirical Approaches : • Focus on 9th grade : Math, French and History teachers and their students • Data : Administrative data matching each teacher to her student (2006-2015) : • Empirical Approaches : • Less than a year absences : exploit variations within teacher, across time, controlling for students’ fixed characteristics (teacher and student fixed effects) • One year or more absences : exploit variations within student, across topics in substitute teachers assignment 5 / 39
Preliminary Results : Less than a Year Absence • Ten days of non replaced days decrease student test scores by 0.6 % of a standard deviation • The number of replaced days on student test scores does not have any statistically significant compensating effect • By type of substitute teacher : • One additional replaced day with a tenured substitute teacher (rather than no class) mitigates 11 % of the marginal impact of absence • One additional replaced day of substitution with a contract teacher (rather than no class) increases the marginal impact of absence by 14 % 6 / 39
Preliminary Results : One Year Absences • Having a contract teacher for a year rather than a regular teacher decreases student test scores by 8.9 % of a standard deviation • Comparable to what would be the yearly effect of the short term absences • Having a tenured substitute teacher for a year rather than a regular teacher decreases student test scores by 1.4 % of a standard deviation 7 / 39
Related Literature • Effect of worker absence on productivity : Miller et al. (2008) ; Clotfelter et al. (2009) ; Duflo et al. (2012) ; Herrmann and Rockoff (2012) • Contract Teachers : Duflo, Dupas and Kremer (2011) • Worker substitutability : Jaravel et al.(2015) ; Hensvik and Rosenqvist (2016) ; J¨ ager(2016) • Instruction time : Pischke (2007) ; Lavy (2015) 8 / 39
Conceptual Framework Institutional Setting Data and Descriptive Statistics Less than a Year Absences One Year Absences Conclusion 9 / 39
Conceptual Framework Institutional Setting Data and Descriptive Statistics Less than a Year Absences One Year Absences Conclusion 10 / 39
Conceptual Framework (1/2) Teacher absences can hurt students through different mechanisms : 1 Loss in instructional time 2 Disruption : • substitute teachers do not have any student-specific human capital • regular teachers do not accumulate student-specific human capital during their absence 3 Difference in ability and experience between the regular and the substitute teachers 11 / 39
Conceptual Framework (2/2) Different potential mechanisms depending on the type of absences : 1 Less than a year absences : • Loss in instructional time • Disruption • Difference of ability/experience between regular and substitute teachers : tenured vs. contract substitute 2 One year of absence : • Difference of ability/experience between regular and substitute teachers : tenured vs. contract substitute 12 / 39
Conceptual Framework Institutional Setting Data and Descriptive Statistics Less than a Year Absences One Year Absences Conclusion 13 / 39
French Middle School : Outline • Teachers are topic-specific • There are no tracks : students keep the same peers for all their different classes • In 9th grade, students take a centralized standardized test score in French, Mathematics and History 14 / 39
Teacher Absence Leave Regulation • Wage setting : • Teachers are civil servants managed centrally by the government • Their wages do not vary across schools and do not depend on output • Teachers cannot be fired • Absence Leave regulation : • Fully paid during their absence leave • No limit in the number of days in paid absence each teacher can take per year (unlike the US) 15 / 39
Types of Absences and Substitute Teachers • Two types of absences : • One year or more : systematically replaced + no disruption for students • Less than a year : not systematically replaced + disruption for students Two types of substitute teachers : • • Tenured teachers assigned to a ZIP code area ( Titulaires sur Zone de Remplacement ) • Contract teachers hired on the spot 16 / 39
Substitute Teachers Hiring • Tenured Substitute Teachers are certified teachers who : • did not get their choice in the centralized assignment procedure to assign teachers (modified version of the deferred acceptance mechanism) • choose to become substitute teachers • Contract teachers : • Conditions : no criminal record and Bachelor’s Degree • Centralized online application • Shortage : educational authority can directly contact people who are registered to the unemployment office 17 / 39
Conceptual Framework Institutional Setting Data and Descriptive Statistics Less than a Year Absences One Year Absences Conclusion 18 / 39
Data Comprehensive administrative data on teachers and students in middle school (2006 - 2015) : • Teachers : • national identifier, school, type of assignment(permanent,yearly, temporary), teaching subject, experience, age, gender • Teachers’ absence and substitution spells : day, month and year of the spells, detailed cause of absence • Middle school students : • encrypted national identifier, socio-demographic characteristics, school, grade and classroom attended • centralized externally graded test scores : DNB (end of 9th grade) 19 / 39
Cumulative Distribution of the Number of Days of Absence per Teacher-Year 1.0 Cumulative share of absent teachers 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 Number of business days per year 20 / 39
Share of Replaced Days by Teacher-Year 1.0 0.9 0.8 0.7 Share of replaced days 0.6 0.5 0.4 0.3 0.2 Substitute Contract Teacher 0.1 Substitute Contract Teacher or Tenured Substitute Teacher 0.0 1 21 41 61 81 101 121 141 161 Number of business days of absence per year 21 / 39
Substitute Teachers Characteristics Regular Teacher. Tenured Sub. Contract Teacher A. Demographics Male 0.36 (0.48) 0.39 (0.49) 0.43 (0.50) Age 43.8 (10.3) 39.0 (10.5) 37.9 (8.9) Average Experience (in years) 14.1 (8.3) 10.0 (8.8) 4.6 (10.2) A year or less of experience 0.02 (0.12) 0.13 (0.34) 0.32 (0.47) B. Certification Agr´ egation 0.05 (0.23) 0.05(0.22) – CAPES 0.77 (0.42) 0.74 (0.44) – Other 0.17 (0.38) 0.21 (0.41) – C. Evaluations Classroom Observation Grade (/60) 46.82(5.99) 44.84 (6.39) 11.85 (9.59) School Principal Grade (/100) 39.02(10.05) 39.15 (11.82) 13.86 (8.70) D. Absences Nb of days of absence 6.58(16.15) 7.48(17.07) 5.53(11.83) Nb of teachers 193,766 67,541 23,035 22 / 39
Contract Teacher Performances at the Certification Exam Contract Teachers Candidates Other Candidates Agreg. CAPES Agreg. CAPES A. Demographics Age (in years) 37.72 (7.75) 35.17 (7.68) 31.05 (8.32) 28.18 (6.65) Male 0.53 (0.50) 0.39 (0.48) 0.46 (0.49) 0.35 (0.48) B. Performance Passing Rate 0.03 (0.18) 0.16 (0.37) 0.15 (0.36) 0.33(0.47) Written Part Grade (/20) 3.91 (2.52) 5.67 (3.14) 6.25 (3.61) 7.30 (3.69) Oral Part Grade (/20) 7.00 (3.78) 7.30 (4.17) 8.09 (3.83) 8.50 (4.58) Nb of obs 286 1,232 8,037 11,779 23 / 39
Proportion of Substitute Teachers by School Percentile Rank at the 9th grade exam 0.12 Tenured Substitute Teacher 0.11 Contract Teacher Proportion of substitute teachers 0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0 20 40 60 80 100 School percentile rank at the 9th grade exam 24 / 39
Conceptual Framework Institutional Setting Data and Descriptive Statistics Less than a Year Absences One Year Absences Conclusion 25 / 39
Recommend
More recommend
